DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The Amendment filed January 9, 2026 has been entered. Claims 1-20 are pending in the application. Applicant has submitted amendments to the claims along with other remarks. Claims 1-20 are still rejected by prior art references, refer to the following rejection for details.
Response to Arguments
Applicant’s arguments and amendments, see pp. 9-12 of the response, filed January 9, 2026, with respect to the rejection(s) of claim(s) 1-20 under § 102 have been fully considered and are persuasive. However, upon further consideration for the amendments, a new ground(s) of rejection is made in view of new reference, please see the rejection for details.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Publication No. 2023/0308199 (hereinafter “Svennebring”) in view of Non-patent Literature entitled, “Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting (hereinafter “Song”).
Regarding claim 1, Svennebring teaches: A device, comprising: a processing system including a processor; and a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the processing system being configured to: categorize users of a cellular network according to a plurality of user categories ([0168] The location information may include, inter alia, the location of specific UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, information about the location of all UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, information about the location of a certain category of UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, a list of UEs 1420 in a particular location, information about the location of all radio nodes currently associated with the MEC server 1336, and/or the like.); identify, by a machine learning model, a service degradation in the cellular network (Abstract, Moreover, the LPP indicates a predicted performance of a radio link between the mobile device and the base station during the future time window. [0190], [0200] In various embodiments, the LPP layer 1502 (or the LPP engine) is configured to perform Multi-Cell Multi-Layer (MCML) data fusion techniques.), wherein the machine learning model comprises: a cell-level model configured to predict a likelihood of service issues at cell sites ([0116] The set of potential transition points represents points in the 3D coordinate space that the mobile device could potentially transition to from the current point. Moreover, the set of potential transition points can include points that are adjacent to the current point (adjacent points) as well as points that are not adjacent to the current point (non-adjacent points).); and a UE-level model configured to identify a source of the service degradation ([0204] The link performance predictions are used to generate the LPP notifications that the LPPS 1500 provides to LPPS consumers such as applications, UEs 1311/1321, and/or NANs 1331-1333, which allow the LPPS consumers to tailor their operations accordingly. The LPP notifications, in one embodiment, are conceptually similar to traffic notifications provided by a navigation application or vehicular driving applications. While the traffic notifications provide up-to-date information about current and forecasted traffic conditions, the LPP notifications provide information about current and forecasted link quality or performance.) by correlating, over a historical time window, temporal features of UE session logs with temporally corresponding predictions from the cell-level model ([0202] The LPP layer 1502 then looks at a time window from a current time to some future time instance (e.g. 30 seconds, 1 minutes, 2 minutes, etc., from the current time depending on how steady the link is), and breaks this time window into time intervals (e.g. 1 second).) identify at least one affected user, the at least one affected user being affected by the service degradation ([0202] The determined LPP for each remaining portions of the time window may be sent to the UE 1311, 1321 or other LPPS consumer, taking into account the amount of delay and predicted location of the UE 1311, 1321 (or other LPPS consumer).); identify one or more affected user categories including the at least one affected user ([0168] The LS supports a location subscribe mechanism, for example, the location is able to be reported multiple times for each location request, periodically or based on specific events, such as location change.); identify potentially affected users, the potentially affected users being categorized according to the one or more affected user categories ([0168] The location services (LS), when available, may provide authorized MEC Apps 1436 with location-related information, and expose such information to the MEC Apps 1436. With location related information, the MEC platform 1437 or one or more MEC Apps 1436 perform active device location tracking, location-based service recommendations, and/or other like services.); and take action to isolate the potentially affected users from the service degradation ([0169] The bandwidth management services (BWMS) provides for the allocation of bandwidth (BW) to certain traffic routed to and from MEC Apps 1436, and specify static/dynamic up/down BW resources, including BW size and BW priority. MEC Apps 1436 may use the BWMS to update/receive BW information to/from the MEC platform 1437. In some embodiments, different MEC Apps 1436 running in parallel on the same MEC server 1336 may be allocated specific static, dynamic up/down BW resources, including BW size and BW priority. The BWMS includes a BW management (BWM) API 1453 to allowed registered applications to statically and/or dynamically register for specific BW allocations per session/application. The BWM API 1453 includes HTTP protocol bindings for BWM functionality using RESTful services or some other suitable API mechanism. [0206] In some embodiments, the LPPS 1500 may be used for efficient small cell backhaul planning. For example, the link performance predictions in the LPP notifications may be used to identify locations to deploy small cells (e.g., RAN nodes 1332, AP 1333, or the like) to enhance coverage and/or capacity while taking into account the associated backhaul and power requirements for such deployments. Additionally or alternatively, in some embodiments, the LPPS 1500 may be used for backhaul link resource allocation. In these embodiments, the link performance predictions may indicate the load or usage of different backhaul links at different time periods on different days, and network operators can reduce operating costs by powering down backhaul links which are expected to require low usage at particular times before powering them back up at predicted peak usage times.).
Svennebring does not teach: wherein the cell-level model comprises: a plurality of spatial-temporal conversion blocks, each block including: (i) a first temporal convolutional layer configured to extract temporal correlations from key performance indicator (KPI) data at neighboring cell sites over a historical time window, (ii) a spatial graph convolutional layer configured to aggregate spatial features of the KPI data from the neighboring cell sites based on a graph-based adjacency matrix and (iii) a second temporal convolutional layer configured to extract temporal features from the aggregated features produced by the spatial graph convolutional layer for each cell site.
However, in the same field of endeavor, Song teaches: wherein the cell-level model comprises: a plurality of spatial-temporal conversion blocks, each block including: (i) a first temporal convolutional layer configured to extract temporal correlations from key performance indicator (KPI) data at neighboring cell sites over a historical time window (e.g., Figure 4, time step 1), (ii) a spatial graph convolutional layer configured to aggregate spatial features of the KPI data from the neighboring cell sites based on a graph-based adjacency matrix (p. 4/8, col. 1, “Graph convolutional operations can aggregate the features of each node with its neighbors.”) and (iii) a second temporal convolutional layer configured to extract temporal features from the aggregated features produced by the spatial graph convolutional layer for each cell site (p. 4/8, col. 2, “We use h(l) to denote the output of the l-th graph convolutional operation, where h(o) denotes the input of the first graph convolutional operation.”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Svennebring to include the feature of spatial or temporal (e.g., spatial/temporal layers) to define a cell-level model and a combination of Svennebring with Song renders the claim prima facie obvious within the described scope of the prior art and any indicated differences within the level of one of ordinary skill in the art (e.g., telecommunications engineer) according to a combination of known prior art elements with known methods to yield predictable results. MPEP 2143(I)(A) (e.g., providing spatial or temporal (e.g., spatial/temporal layers) to define a cell-level model).
Regarding claim 2, Svennebring teaches: identify network protections isolating a potentially affected user of the potentially affected users from the service degradation; and take no further action to isolate the potentially affected users based on the network protections ([0210] Additionally or alternatively, content may be pre-fetched an loaded to CDN edge compute nodes 1330 along a predicted travel path/route of a UE 1321 (e.g., closer to the UE’s 1321 anticipated point of consumption), and the amount of content streamed to the UE 1321 at different points along the predicted path/route may be based on predicted network performance at those different points.).
Regarding claim 3, Svennebring teaches: identify a type of service degradation in the cellular network; identify user vulnerability to the type of service degradation; and group in a same category users having a same user vulnerability to the type of service degradation ([0213] In a fourth example, a mobile network operator (e.g., an owner/operator of CN 1342 and/or some or all of the NANs 1331, 1332) uses to guarantee a certain level of network performance (e.g., Service Level Guarantees) for enterprise-level customers. In this example, the mobile network operator may use the LPP notifications and/or the LPPS 1500 to quantify their network performance at varying levels of granularities showing their customers that they can provide a consistent high-quality QoS/QoE.).
Regarding claim 4, Svennebring teaches: identify an application operated by a user on a user equipment device (UE device) of the user; and group in a same category users operating a same application a UE device of the user ([0211] In a second example, a service provider provides over-the-top (OTT) real-time services including, for example, television, messaging (e.g., instant messaging, online chats, etc.), and voice calling. In this example, the service provider uses the LPP notifications to provide advance warnings to users indicating when an OTT connection will likely be lost, and how long the connection may take to be reconnected.).
Regarding claim 5, Svennebring teaches: identify a quality of service (QoS) class identifier (QCI) for the user and the application; and group in a same category users having a same QCI ([0169] The BWMS includes a BW management (BWM) API 1453 to allowed registered applications to statically and/or dynamically register for specific BW allocations per session/application.).
Regarding claim 6, Svennebring teaches: identify a location of a user; and group in a same category users based on the location of the user ([0168] The Zonal Presence service utilizes the concept of “zone”, where a zone lends itself to be used to group all radio nodes that are associated to a MEC host or MEC server 1336, or a subset thereof, according to a desired deployment. In this regard, the OMA Zonal Presence API 1453 provides means for MEC Apps 1436 to retrieve information about a zone, the access points associated to the zones and the users that are connected to the access points.).
Regarding claim 7, Svennebring teaches: identify one of a geographic location of the user and a network location of the user ([0168] The location information may be in the form of a geolocation, a Global Navigation Satellite Service (GNSS) coordinate, a Cell identity (ID), and/or the like.).
Regarding claim 8, Svennebring teaches: receive, from the machine learning model, information about a network problem probability ([0037] Based on the various types of data 132a-e provided as input (e.g., historical, real-time, and feedback data), the LPP server 130 can predict future link performance for client devices with a high level of confidence and accuracy. For example, the LPP server 130 may generate a prediction 135 regarding the future performance of a network link 115 to a particular client device 110. In some embodiments, a prediction 135 generated by the LPP server 130 may include or indicate a time (e.g., time at which the predicted behavior will occur), a type of prediction (e.g., bandwidth or latency prediction), a predicted value (e.g., predicted amount of bandwidth or latency), an expected deviation (e.g., expected amount of deviation from the predicted bandwidth or latency value), and/or a probability (e.g., the likelihood or confidence of the prediction being correct), among other types of information.); and identify the action to isolate the potentially affected users from the service degradation based on the network problem probability ([0123] In some cases, the recipient of the LPP may take certain actions in response to the predicted link performance, such as proactively compensating for an expected signal quality reduction and/or taking advantage of an upcoming high-speed coverage area. In response to an expected reduction in link performance, for example, certain proactive measures may be taken, such as performing a handoff to another base station, re-routing the mobile device through a better coverage area, adjusting video/media streaming parameters (e.g., resolution, frame rate, pre-buffer length), and so forth.).
Regarding claim 9, Svennebring teaches: hand off communication between a user equipment device of a user from a first cell site to a second site, wherein the first cell site is affected by the service degradation ([0049] For example, in coverage scenario 200b, a connection is maintained with one of two base stations 202a,b during the entire path 205a,b of the UE 204 (as depicted by the solid line 205a,b). By providing a hint to the UE 204 about the possibility of an unstable connection with base station 202a at time t.sub.2, a handoff can be performed before time t.sub.2-at time t.sub.H-to handover the UE 204 from base station 202a to base station 202b.).
Regarding claim 10, Svennebring teaches: receive, from the UE level model, information identifying a source of the service degradation in the cellular network ([0204] the LPP notifications provide information about current and forecasted link quality or performance).
Regarding claim 11, Svennebring teaches: A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the processing system being configured to: receive information identifying a user of a user equipment device (UE device) in a cellular network ([0201] In various embodiments, MCML data fusion accounts for different operational states, operational contexts, and/or mobility states. For example, as discussed in more detail infra, a cell transition prediction layer may be used to predict the cells that a UE 1311, 1321 will visit at particular time instances based on a current cell in which the UE 1311, 1321 is camping, a previous cell visited by the UE 1311, 1321, and travel direction and velocity measurements.); group the user with a plurality of other users in categories, wherein the grouping is responsive to an activity or a location of the user and the plurality of other users ([0168] The location information may include, inter alia, the location of specific UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, information about the location of all UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, information about the location of a certain category of UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, a list of UEs 1420 in a particular location, information about the location of all radio nodes currently associated with the MEC server 1336, and/or the like.); receive, from a machine learning model, information about a network failure in the cellular network (Abstract, Moreover, the LPP indicates a predicted performance of a radio link between the mobile device and the base station during the future time window. [0200] In various embodiments, the LPP layer 1502 (or the LPP engine) is configured to perform Multi-Cell Multi-Layer (MCML) data fusion techniques.), wherein the machine learning model comprises; a cell-level model configured to predict a likelihood of service issues at cell sites ([0116] The set of potential transition points represents points in the 3D coordinate space that the mobile device could potentially transition to from the current point. Moreover, the set of potential transition points can include points that are adjacent to the current point (adjacent points) as well as points that are not adjacent to the current point (non-adjacent points).); and a UE-level model configured to identify a source of the network failure ([0204] The link performance predictions are used to generate the LPP notifications that the LPPS 1500 provides to LPPS consumers such as applications, UEs 1311/1321, and/or NANs 1331-1333, which allow the LPPS consumers to tailor their operations accordingly. The LPP notifications, in one embodiment, are conceptually similar to traffic notifications provided by a navigation application or vehicular driving applications. While the traffic notifications provide up-to-date information about current and forecasted traffic conditions, the LPP notifications provide information about current and forecasted link quality or performance.) by correlating, over a historical time window, temporal features of UE session logs with temporally corresponding predictions from the cell-level model ([0202] The LPP layer 1502 then looks at a time window from a current time to some future time instance (e.g. 30 seconds, 1 minutes, 2 minutes, etc., from the current time depending on how steady the link is), and breaks this time window into time intervals (e.g. 1 second).) identify at least one affected user, the at least one affected user being affected by the service degradation ([0202] The determined LPP for each remaining portions of the time window may be sent to the UE 1311, 1321 or other LPPS consumer, taking into account the amount of delay and predicted location of the UE 1311, 1321 (or other LPPS consumer).); identify one or more affected user categories including the at least one affected user ([0168] The LS supports a location subscribe mechanism, for example, the location is able to be reported multiple times for each location request, periodically or based on specific events, such as location change.); identify potentially affected users, the potentially affected users being categorized according to the one or more affected user categories ([0168] The location services (LS), when available, may provide authorized MEC Apps 1436 with location-related information, and expose such information to the MEC Apps 1436. With location related information, the MEC platform 1437 or one or more MEC Apps 1436 perform active device location tracking, location-based service recommendations, and/or other like services.); and take action to isolate the potentially affected users from the service degradation ([0169] The bandwidth management services (BWMS) provides for the allocation of bandwidth (BW) to certain traffic routed to and from MEC Apps 1436, and specify static/dynamic up/down BW resources, including BW size and BW priority. MEC Apps 1436 may use the BWMS to update/receive BW information to/from the MEC platform 1437. In some embodiments, different MEC Apps 1436 running in parallel on the same MEC server 1336 may be allocated specific static, dynamic up/down BW resources, including BW size and BW priority. The BWMS includes a BW management (BWM) API 1453 to allowed registered applications to statically and/or dynamically register for specific BW allocations per session/application. The BWM API 1453 includes HTTP protocol bindings for BWM functionality using RESTful services or some other suitable API mechanism. [0206] In some embodiments, the LPPS 1500 may be used for efficient small cell backhaul planning. For example, the link performance predictions in the LPP notifications may be used to identify locations to deploy small cells (e.g., RAN nodes 1332, AP 1333, or the like) to enhance coverage and/or capacity while taking into account the associated backhaul and power requirements for such deployments. Additionally or alternatively, in some embodiments, the LPPS 1500 may be used for backhaul link resource allocation. In these embodiments, the link performance predictions may indicate the load or usage of different backhaul links at different time periods on different days, and network operators can reduce operating costs by powering down backhaul links which are expected to require low usage at particular times before powering them back up at predicted peak usage times.).
Svennebring does not teach: wherein the cell-level model comprises: a plurality of spatial-temporal conversion blocks, each block including: (i) a first temporal convolutional layer configured to extract temporal correlations from key performance indicator (KPI) data at neighboring cell sites over a historical time window, (ii) a spatial graph convolutional layer configured to aggregate spatial features of the KPI data from the neighboring cell sites based on a graph-based adjacency matrix and (iii) a second temporal convolutional layer configured to extract temporal features from the aggregated features produced by the spatial graph convolutional layer for each cell site.
However, in the same field of endeavor, Song teaches: wherein the cell-level model comprises: a plurality of spatial-temporal conversion blocks, each block including: (i) a first temporal convolutional layer configured to extract temporal correlations from key performance indicator (KPI) data at neighboring cell sites over a historical time window (e.g., Figure 4, time step 1), (ii) a spatial graph convolutional layer configured to aggregate spatial features of the KPI data from the neighboring cell sites based on a graph-based adjacency matrix (p. 4/8, col. 1, “Graph convolutional operations can aggregate the features of each node with its neighbors.”) and (iii) a second temporal convolutional layer configured to extract temporal features from the aggregated features produced by the spatial graph convolutional layer for each cell site (p. 4/8, col. 2, “We use h(l) to denote the output of the l-th graph convolutional operation, where h(o) denotes the input of the first graph convolutional operation.”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Svennebring to include the feature of spatial or temporal (e.g., spatial/temporal layers) to define a cell-level model and a combination of Svennebring with Song renders the claim prima facie obvious within the described scope of the prior art and any indicated differences within the level of one of ordinary skill in the art (e.g., telecommunications engineer) according to a combination of known prior art elements with known methods to yield predictable results. MPEP 2143(I)(A) (e.g., providing spatial or temporal (e.g., spatial/temporal layers) to define a cell-level model).
Regarding claim 12, Svennebring teaches: receive, from the machine learning model, information identifying a source of the network failure in the cellular network ([0204] The link performance predictions are used to generate the LPP notifications that the LPPS 1500 provides to LPPS consumers such as applications, UEs 1311/1321, and/or NANs 1331-1333, which allow the LPPS consumers to tailor their operations accordingly. The LPP notifications, in one embodiment, are conceptually similar to traffic notifications provided by a navigation application or vehicular driving applications. While the traffic notifications provide up-to-date information about current and forecasted traffic conditions, the LPP notifications provide information about current and forecasted link quality or performance.).
Regarding claim 13, Svennebring teaches: identify a type of network failure in the cellular network; identify user vulnerability to the type of network failure; and group in a same category users having a same user vulnerability to the type of network failure ([0213] In a fourth example, a mobile network operator (e.g., an owner/operator of CN 1342 and/or some or all of the NANs 1331, 1332) uses to guarantee a certain level of network performance (e.g., Service Level Guarantees) for enterprise-level customers. In this example, the mobile network operator may use the LPP notifications and/or the LPPS 1500 to quantify their network performance at varying levels of granularities showing their customers that they can provide a consistent high-quality QoS/QoE.).
Regarding claim 14, Svennebring teaches: receive, from the machine learning model, information about a network problem probability ([0037] Based on the various types of data 132a-e provided as input (e.g., historical, real-time, and feedback data), the LPP server 130 can predict future link performance for client devices with a high level of confidence and accuracy. For example, the LPP server 130 may generate a prediction 135 regarding the future performance of a network link 115 to a particular client device 110. In some embodiments, a prediction 135 generated by the LPP server 130 may include or indicate a time (e.g., time at which the predicted behavior will occur), a type of prediction (e.g., bandwidth or latency prediction), a predicted value (e.g., predicted amount of bandwidth or latency), an expected deviation (e.g., expected amount of deviation from the predicted bandwidth or latency value), and/or a probability (e.g., the likelihood or confidence of the prediction being correct), among other types of information.); and identify the action to isolate the potentially affected users from the service degradation based on the network problem probability ([0123] In some cases, the recipient of the LPP may take certain actions in response to the predicted link performance, such as proactively compensating for an expected signal quality reduction and/or taking advantage of an upcoming high-speed coverage area. In response to an expected reduction in link performance, for example, certain proactive measures may be taken, such as performing a handoff to another base station, re-routing the mobile device through a better coverage area, adjusting video/media streaming parameters (e.g., resolution, frame rate, pre-buffer length), and so forth.).
Regarding claim 15, Svennebring teaches: hand off communication between the UE device and the first cell site to a second cell site, wherein handing off communication is based on the information about the risk of the network failure at a first cell site ([0049] For example, in coverage scenario 200b, a connection is maintained with one of two base stations 202a,b during the entire path 205a,b of the UE 204 (as depicted by the solid line 205a,b). By providing a hint to the UE 204 about the possibility of an unstable connection with base station 202a at time t.sub.2, a handoff can be performed before time t.sub.2-at time t.sub.H-to handover the UE 204 from base station 202a to base station 202b.).
Regarding claim 16, Svennebring teaches: A method, comprising: receiving, by a processing system including a processor, information identifying a user of a user equipment device (UE device) in a cellular network ([0201] In various embodiments, MCML data fusion accounts for different operational states, operational contexts, and/or mobility states. For example, as discussed in more detail infra, a cell transition prediction layer may be used to predict the cells that a UE 1311, 1321 will visit at particular time instances based on a current cell in which the UE 1311, 1321 is camping, a previous cell visited by the UE 1311, 1321, and travel direction and velocity measurements.); grouping, by the processing system, a user associated with the UE device with a plurality of other users in the cellular network in a plurality of categories, wherein the grouping is based on common features of the user and the plurality of other users ([0168] The location information may include, inter alia, the location of specific UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, information about the location of all UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, information about the location of a certain category of UEs 1420 currently served by the radio node(s) associated with the MEC server 1336, a list of UEs 1420 in a particular location, information about the location of all radio nodes currently associated with the MEC server 1336, and/or the like.); receiving, by the processing system, from a machine learning model, information about a service degradation in the cellular network, (Abstract, Moreover, the LPP indicates a predicted performance of a radio link between the mobile device and the base station during the future time window. [0200] In various embodiments, the LPP layer 1502 (or the LPP engine) is configured to perform Multi-Cell Multi-Layer (MCML) data fusion techniques.), wherein the machine learning model comprises; a cell-level model configured to predict a likelihood of service issues at cell sites ([0116] The set of potential transition points represents points in the 3D coordinate space that the mobile device could potentially transition to from the current point. Moreover, the set of potential transition points can include points that are adjacent to the current point (adjacent points) as well as points that are not adjacent to the current point (non-adjacent points).); and a UE-level model configured to identify a source of the network failure ([0204] The link performance predictions are used to generate the LPP notifications that the LPPS 1500 provides to LPPS consumers such as applications, UEs 1311/1321, and/or NANs 1331-1333, which allow the LPPS consumers to tailor their operations accordingly. The LPP notifications, in one embodiment, are conceptually similar to traffic notifications provided by a navigation application or vehicular driving applications. While the traffic notifications provide up-to-date information about current and forecasted traffic conditions, the LPP notifications provide information about current and forecasted link quality or performance.) by correlating, over a historical time window, temporal features of UE session logs with temporally corresponding predictions from the cell-level model ([0202] The LPP layer 1502 then looks at a time window from a current time to some future time instance (e.g. 30 seconds, 1 minutes, 2 minutes, etc., from the current time depending on how steady the link is), and breaks this time window into time intervals (e.g. 1 second).) identify at least one affected user, the at least one affected user being affected by the service degradation ([0202] The determined LPP for each remaining portions of the time window may be sent to the UE 1311, 1321 or other LPPS consumer, taking into account the amount of delay and predicted location of the UE 1311, 1321 (or other LPPS consumer).); identify one or more affected user categories including the at least one affected user ([0168] The LS supports a location subscribe mechanism, for example, the location is able to be reported multiple times for each location request, periodically or based on specific events, such as location change.); identify potentially affected users, the potentially affected users being categorized according to the one or more affected user categories ([0168] The location services (LS), when available, may provide authorized MEC Apps 1436 with location-related information, and expose such information to the MEC Apps 1436. With location related information, the MEC platform 1437 or one or more MEC Apps 1436 perform active device location tracking, location-based service recommendations, and/or other like services.); and take action to isolate the potentially affected users from the service degradation ([0169] The bandwidth management services (BWMS) provides for the allocation of bandwidth (BW) to certain traffic routed to and from MEC Apps 1436, and specify static/dynamic up/down BW resources, including BW size and BW priority. MEC Apps 1436 may use the BWMS to update/receive BW information to/from the MEC platform 1437. In some embodiments, different MEC Apps 1436 running in parallel on the same MEC server 1336 may be allocated specific static, dynamic up/down BW resources, including BW size and BW priority. The BWMS includes a BW management (BWM) API 1453 to allowed registered applications to statically and/or dynamically register for specific BW allocations per session/application. The BWM API 1453 includes HTTP protocol bindings for BWM functionality using RESTful services or some other suitable API mechanism. [0206] In some embodiments, the LPPS 1500 may be used for efficient small cell backhaul planning. For example, the link performance predictions in the LPP notifications may be used to identify locations to deploy small cells (e.g., RAN nodes 1332, AP 1333, or the like) to enhance coverage and/or capacity while taking into account the associated backhaul and power requirements for such deployments. Additionally or alternatively, in some embodiments, the LPPS 1500 may be used for backhaul link resource allocation. In these embodiments, the link performance predictions may indicate the load or usage of different backhaul links at different time periods on different days, and network operators can reduce operating costs by powering down backhaul links which are expected to require low usage at particular times before powering them back up at predicted peak usage times.).
Svennebring does not teach: wherein the cell-level model comprises: a plurality of spatial-temporal conversion blocks, each block including: (i) a first temporal convolutional layer configured to extract temporal correlations from key performance indicator (KPI) data at neighboring cell sites over a historical time window, (ii) a spatial graph convolutional layer configured to aggregate spatial features of the KPI data from the neighboring cell sites based on a graph-based adjacency matrix and (iii) a second temporal convolutional layer configured to extract temporal features from the aggregated features produced by the spatial graph convolutional layer for each cell site.
However, in the same field of endeavor, Song teaches: wherein the cell-level model comprises: a plurality of spatial-temporal conversion blocks, each block including: (i) a first temporal convolutional layer configured to extract temporal correlations from key performance indicator (KPI) data at neighboring cell sites over a historical time window (e.g., Figure 4, time step 1), (ii) a spatial graph convolutional layer configured to aggregate spatial features of the KPI data from the neighboring cell sites based on a graph-based adjacency matrix (p. 4/8, col. 1, “Graph convolutional operations can aggregate the features of each node with its neighbors.”) and (iii) a second temporal convolutional layer configured to extract temporal features from the aggregated features produced by the spatial graph convolutional layer for each cell site (p. 4/8, col. 2, “We use h(l) to denote the output of the l-th graph convolutional operation, where h(o) denotes the input of the first graph convolutional operation.”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Svennebring to include the feature of spatial or temporal (e.g., spatial/temporal layers) to define a cell-level model and a combination of Svennebring with Song renders the claim prima facie obvious within the described scope of the prior art and any indicated differences within the level of one of ordinary skill in the art (e.g., telecommunications engineer) according to a combination of known prior art elements with known methods to yield predictable results. MPEP 2143(I)(A) (e.g., providing spatial or temporal (e.g., spatial/temporal layers) to define a cell-level model).
Regarding claim 17, Svennebring teaches: identifying, by the processing system, a type of service degradation in the cellular network; identifying, by the processing system, user vulnerability of the user and other users of a common group to the type of service degradation; and grouping, by the processing system, in a same category, users having a same user vulnerability to the type of service degradation ([0213] In a fourth example, a mobile network operator (e.g., an owner/operator of CN 1342 and/or some or all of the NANs 1331, 1332) uses to guarantee a certain level of network performance (e.g., Service Level Guarantees) for enterprise-level customers. In this example, the mobile network operator may use the LPP notifications and/or the LPPS 1500 to quantify their network performance at varying levels of granularities showing their customers that they can provide a consistent high-quality QoS/QoE.).
Regarding claim 18, Svennebring teaches: grouping, by the processing system, the user and other users based on a commonly used application accessed over the cellular network ([0211] In a second example, a service provider provides over-the-top (OTT) real-time services including, for example, television, messaging (e.g., instant messaging, online chats, etc.), and voice calling. In this example, the service provider uses the LPP notifications to provide advance warnings to users indicating when an OTT connection will likely be lost, and how long the connection may take to be reconnected.).
Regarding claim 19, Svennebring teaches: grouping by the processing system, in a same category users having a same quality of service (QoS) class identifier (QCI) ([0169] The BWMS includes a BW management (BWM) API 1453 to allowed registered applications to statically and/or dynamically register for specific BW allocations per session/application.).
Regarding claim 20, Svennebring teaches: grouping, by the processing system, in a same category users having a same location ([0168] The Zonal Presence service utilizes the concept of “zone”, where a zone lends itself to be used to group all radio nodes that are associated to a MEC host or MEC server 1336, or a subset thereof, according to a desired deployment. In this regard, the OMA Zonal Presence API 1453 provides means for MEC Apps 1436 to retrieve information about a zone, the access points associated to the zones and the users that are connected to the access points.).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JUSTIN BARRY whose telephone number is (571)272-0201. The examiner can normally be reached 8:00am EST to 5:00pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jinsong HU can be reached at (571) 272-3965. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JAB/ Examiner, Art Unit 2643
/JINSONG HU/ Supervisory Patent Examiner, Art Unit 2643